8 replies to this topic

[Vyoma]

Hi,

Currently, as for past 50-60 minutes, all MySQL databases are down. Ticket ID: NRY-328984

Normally - I do not open a thread here when there is a problem. I just raise a ticket. But there is a cause of concern for me at this point in time.
One thing is that I have been escalating the status of the ticket and updating with my findings for the past duration since it was down - there has been no response - not a single word from the support team.

Second, the number of times the databases are down. I went back and checked up my notification setup and these are the dates:
09-Jan-2010 - Ticket ID: WWH-960267
21-Jan-2010 - No ticket but issue occured when I was not checking. But I have notification email sent from vBulletin installation that the databases was down.
22-Jan-2010 - This is the current issue - ongoing that has not been fixed as of now. I am still receiving notification emails that the databases are down.

Out of the set of sites I run, one of them has pretty high traffic and earns revenue round the clock. Another one, I run website optimization experiments and conversion tests which assume that the websites are functional throughout the testing duration. The websites being down since the databases were not working is throwing a wrench into all the effort.

Why are not the MySQL databases not monitored and alerts set for the support personnel?

[shadowx]

http://www.xisto.com/network-status/

You can monitor the server status from that link. I cant help any more than that as i too am simply a hosted member. You must be on the Epsilon server as it is the only one with MYSQL flagged as down.

[Vyoma]

shadowx, on Jan 21 2010, 04:20 PM, said:

http://www.xisto.com/network-status/

You can monitor the server status from that link. I cant help any more than that as i too am simply a hosted member. You must be on the Epsilon server as it is the only one with MYSQL flagged as down.

Well, it is up now 8 hours after the issue began. And the link that you provide does not list the server on which I am hosted. I am checking the cPanel now and it says:
Server Name iota

[anwiii]

the database may be up, but did you ever get a reply to your support ticket? i'm curious why the database was down. i went to two of your drupal sites and both were down due to database issues so i am assuming it probably wasn't on your end. sometimes sites can go down like that and has nothing to do with the server.

glad everything is working for you now. i know it always sucks if you lose any money over it. i have never experienced downtime like that....but according to the 99.9 rule, this allows for 10 more days of 8 hours downtime. it's still a very good percentage of uptime.

if you ever do get a reply, quote it here. i am curious about the answer.

Vyoma, on Jan 22 2010, 01:01 AM, said:

Well, it is up now 8 hours after the issue began. And the link that you provide does not list the server on which I am hosted. I am checking the cPanel now and it says:
Server Name iota

[Vyoma]

anwiii, on Jan 22 2010, 01:14 AM, said:

the database may be up, but did you ever get a reply to your support ticket? i'm curious why the database was down. i went to two of your drupal sites and both were down due to database issues so i am assuming it probably wasn't on your end. sometimes sites can go down like that and has nothing to do with the server.

glad everything is working for you now. i know it always sucks if you lose any money over it. i have never experienced downtime like that....but according to the 99.9 rule, this allows for 10 more days of 8 hours downtime. it's still a very good percentage of uptime.

if you ever do get a reply, quote it here. i am curious about the answer.

Oh yes. They did respond to the support ticket after 9 hours. They did say that their MySQL database was having issues and they have sent it to "senior admin to get a permanant fix". Meanwhile, they are suggesting I move my sites to some "new servers". I am not sure how that is going to help. If they can ensure that MySQL databases remain up and running there, why cant they do the same on the IOTA server?

The 99.9% uptime rule gives them 87.6 hours in a whole year. In a month, that would be 7.3 hours. They have already crossed past the SLA for this month, in my opinion. It is around 9 hours of downtime for this issue. Counting the two other issues that occurred this month, the downtime hours come to 15-16 hours. That in my opinion is not a 99.9% uptime this month. The month is not even over yet, there are 8 more days left.

[anwiii]

so they said it it was only fixed temporarily and not permanantly and now suggesting you move your entire sites? hmmm, that is some strange advice there to move your entire sites. maybe you need to speak with an admin directly on this issue. they may be able to enlighten you more on this issue.

i know when we create our websites, they are created to be seen and downtime should never be an excuse. i know if i had a website and experienced more downtime than i was willing to accept, i would change hosts. but changing hosts and servers create a problem in itself when the original problem(databases in this instance) can be fixed before you switch hosts or servers. but that is your other option. that's strange why that solution would be suggested to you though. that seems like it should be more of a personal choice. not a suggestion coming from support.

so i would really talk to an admin on this issue. request that an admin handle your ticket and your future concerns about downtime. your websites seem to have been online for a while according to all the content that has been posted on them and this is the first time i've heard you complain about downtime...so maybe if we compare how long you've been hosted to how much downtime there has been, they still fall within the 99.9.

you are complaining about the downtime this month...which you have every right to. i would be upset too....but your website wasn't created this month and has been around a lot longer. who knows....maybe they have a better track record overall than 99.90. ya think?

Vyoma, on Jan 22 2010, 02:32 AM, said:

Oh yes. They did respond to the support ticket after 9 hours. They did say that their MySQL database was having issues and they have sent it to "senior admin to get a permanant fix". Meanwhile, they are suggesting I move my sites to some "new servers". I am not sure how that is going to help. If they can ensure that MySQL databases remain up and running there, why cant they do the same on the IOTA server?

The 99.9% uptime rule gives them 87.6 hours in a whole year. In a month, that would be 7.3 hours. They have already crossed past the SLA for this month, in my opinion. It is around 9 hours of downtime for this issue. Counting the two other issues that occurred this month, the downtime hours come to 15-16 hours. That in my opinion is not a 99.9% uptime this month. The month is not even over yet, there are 8 more days left.

[shadowx]

I would follow their advice to switch servers. Evidently your server has some stability issues.

Unless all of the servers are exact copies of each other then some will perform better than others. And even if the software is cloned the user added content makes problems harder to track down.

I understand your frustration, i would feel the same way but the fix is there, it is just up to you to take it.

If you arent sure how to switch servers add it to your ticket and ask for a guide.

[Vyoma]

shadowx, on Jan 22 2010, 02:32 AM, said:

I would follow their advice to switch servers. Evidently your server has some stability issues.

Unless all of the servers are exact copies of each other then some will perform better than others. And even if the software is cloned the user added content makes problems harder to track down.

I understand your frustration, i would feel the same way but the fix is there, it is just up to you to take it.

If you arent sure how to switch servers add it to your ticket and ask for a guide.

Yes shadowx. I figured that would be the best approach after pondering about it for a while and asked for them to switch to a newer server in the ticket.

anwiii, on Jan 22 2010, 01:58 AM, said:

so they said it it was only fixed temporarily and not permanantly and now suggesting you move your entire sites? hmmm, that is some strange advice there to move your entire sites. maybe you need to speak with an admin directly on this issue. they may be able to enlighten you more on this issue.

i know when we create our websites, they are created to be seen and downtime should never be an excuse. i know if i had a website and experienced more downtime than i was willing to accept, i would change hosts. but changing hosts and servers create a problem in itself when the original problem(databases in this instance) can be fixed before you switch hosts or servers. but that is your other option. that's strange why that solution would be suggested to you though. that seems like it should be more of a personal choice. not a suggestion coming from support.

so i would really talk to an admin on this issue. request that an admin handle your ticket and your future concerns about downtime. your websites seem to have been online for a while according to all the content that has been posted on them and this is the first time i've heard you complain about downtime...so maybe if we compare how long you've been hosted to how much downtime there has been, they still fall within the 99.9.

you are complaining about the downtime this month...which you have every right to. i would be upset too....but your website wasn't created this month and has been around a lot longer. who knows....maybe they have a better track record overall than 99.90. ya think?

Oh yes, anwiii, my sites have been up and running for a long time. I had been a member of AstaHost for couple of months, hosted through credits that I earned from posting there. After a couple of months, when my website picked up, I notice that it was having repeated downtime. But I did not complain then, other than issuing a ticket to get it fixed. Since the hosting was provided for the effort I put into posting relevant content at AstaHost, I realized that I could not complain about that.

After that, I switched to paid hosting over here at ComputingHost.com and moved over my website. I even grew the number of websites I have been hosting. I started with a 3S plan. And it did have downtime issues - and then I complained. They fixed them as and when the issues came up, and once I did see them slack up on meeting their SLA. After a year or so, I switched to a 7S plan since I had a pretty large website and was highly trafficed ( KalaaLog.com ). When I ran into issues, I did raise tickets and follow up with them. I have around 30 support tickets raised from 2007 to current date. Of course, couple of them were sales related questions or minor cron related issues. But the rest of them were either problems with DNS or MySQL databases being down - critical issues. When this happens, my websites are not accesible.

So speaking overall uptime, it might be 99.9% but that does not mean for a service period they can have a downtime. I understand that asking for a 99.9% uptime every day (which is around 14 minutes downtime) is not practical. I am not asking for that. (I know since I too have a day job where I support applications and systems that adhere to SLA uptimes). But for a duration of one month, and a subscription period of 6 months, if downtime is 40 odd hours, I will be very very concerned about the remaining five months. That is the reason, I bring this issue up when they have gone beyond the 99.9% uptime for a month (and hence the title of the thread says that the database was down for three times in January).

The complaints I raise are from the perspecitve that I want the Xisto group to do well and not with an malicious intent. I owe some of my loyalty to AstaHost and its sister sites, since that is where I started and they supported me.

Basically, I am lazy. I do not want to go out searching for another reliable webhost. If I can raise a complain, and get ComputingHost/Xisto to do better then it is good for me. I would like to grow my websites to an extent where I would need to take up VPS Hosting from them, may be even Dedicated Hosting at ComputingHost one day. But to do that, I would want to have a sustained growth and that can be only done when my current hosting is as reliable as advertized on the homepage. If ComputingHost/Xisto does not do good, it is not good for me. I will need to re-evaluate other services and that is a costly step in terms of effort for me.

Ooo boy. I turned out to be big rant.

Edited by Vyoma, 22 January 2010 - 03:03 PM.

[mahesh2k]

I faced this database issue during period 22 December to 12th January. One of my site on Kappa and theta ( i don't remember exact names, have to open tickets for that) were down during this period. I think traffic spike sometimes makes it hard on server. I had huge spike from stumbleupon and entrecard during this period so yes, this might be issue or there is likely to be some issue with the Mysql server.